| Abstract: |
In recent years, open science and metascience have made advances by reducing
publication bias, applying largeteam science, and addressing reproducibility.
Although both approaches are increasingly being incorporated into many
scientific fields, it is unclear to what extent they are appreciated by the
public. We evaluate the public’s perception of scientific findings by exposing
more than 7, 500 participants from seven countries that vary in general trust
in science. In this randomized experiment, participants are exposed to
versions of a similar scientific result pointing toward the same conclusion:
participants read about a small classical single-team study, a large sample
size study, or a many-labs (many-teams) study with 21 research teams working
independently. We find that in aggregate, the public considers both the
many-labs and the large-sample study more credible, generalizable and relevant
for society than the single-team study. Importantly, they see the involvement
of many research teams and having many study participants as key advantages.
Moreover, we find no statistically significant differences in perceptions of
the many-labs study and the large-sample study. We also find similar results
within some countries and null results within other countries, emphasizing the
importance of studying these questions in different contexts. These findings
underscore the importance of novel metascience and large-sample studies to
improve the credibility of scientific research in the general population. It
also underscores the increasing importance of communicating the benefits and
findings of metascience and large-sample projects to the public. |